Study on plasmatic metabolomics of Uygur patients with essential hypertension based on nuclear magnetic resonance technique

Abstract

Objective: Metabolomics is the analysis of the global constitution of endogenous metabolites in cells, tissue, and bodily fluids based on analysis techniques with high output, high sensitivity, and high resolution. The physiological and pathological state of the subject investigated could be identified and analyzed through examining metabolite changes. In this study, 1H-NMR metabolomics was employed to study plasma metabolites of both Uygur patients with hypertension and healthy people, thereby filtering out characteristic metabolites for Uygur patients with hypertension. The pathogenesis of hypertension was discussed via metabolic pathways.

Patients and methods: A total of 256 Uygur subjects were recruited for this study and divided into two groups, namely hypertension group (157 Uygur patients with hypertension) and normal group (99 healthy Uygur subjects). They were all taken from epidemiological surveys on the Uygur people of Qira County, Hotan, and Xinjiang between 2009 and 2012 conducted by Hypertension Research Group of Xinjiang Medical University. Furthermore, all subjects have Uygur blood within three generations. For the hypertension group, the recruitment criteria is systolic blood pressure (SBP) ≥ 140 mmHg (lmmHg = 0.133 kPa) and/or diastolic blood pressure (DBP) ≥ 90 mmHg at rest. Patients who had taken antihypertensive drugs within two weeks and those who were diagnosed with essential hypertension (EH) were also included in this group, while patients with secondary hypertension, myocardiosis, congenital heart disease, and rheumatic valvular heart disease were not included. In the healthy normal-pressure group, blood pressures were within normal range: SBP < 140 mmHg, DBP < 90 mmHg, without history of antihypertensive drugs, cardiovascular and cerebrovascular diseases, and liver/kidney diseases. All subjects were measured separately with Inova600 nuclear magnetic resonance (NMR) spectrometer to conduct the 1H-NMR experiment. Serum specimens from both the hypertension group and the healthy control group were used for NMR spectrograms before data pre-processing, where aggregate analysis was performed for NMR data/metabolic information with principal component analysis (PCA). Then, partial least squares discriminant analysis (PLS-DA) was employed to classify and predict different groups of specimens, and orthogonal partial least squares discriminant analysis (OPLS-DA) was conducted to cross-validate the quality of the models. Statistical analysis was further performed to test significance of the correlation coefficient to determine differential metabolic components in serums of both groups of subjects. Based on the information from the differential metabolic components, a metabolic pathway network related to hypertension could be constructed, thereby revealing potential biomarkers for hypertension.

Results: Clinical data showed that subjects in the two groups were not significantly different with respect to age, weight, and height, as well as lipid indices, including TG, LDL (p > 0.05), while FPG, SBP, DBP, HDL, and TC were significantly different between the two groups (p < 0.05). OPLS-DA results demonstrated that integral quantities of principal components were mainly distributed within four areas of the ellipse scatter diagram (95% confidence interval). From the score plot and 3D distribution diagram, it can be observed that the distribution areas for the two groups are completely separate, thereby indicating that the serum of the Uygur hypertension patients is significantly different from that of healthy subjects in terms of metabolic components. OPLS-DA results indicate that differences in metabolic components are significant between the two groups, and 12 different metabolites were identified. Compared to healthy subjects, patients with hypertension possess a much lower quantity of many amino acids, including valine, alanine, pyroracemic acid, inose, p-hydroxyphenylalanine, and methylhistidine, among others (p < 0.05), with a significant increase in VLDL, LDL, lactic acid, and acetone (p < 0.05).

Conclusions: The 1H-NMR metabolomics process, in combination with OPLS-DA pattern identification, is an effective way to differentiate the serum metabolites characteristic of hypertension patients. Pattern identification analysis of NMR spectrum data with OPLS-DA could identify metabolites of hypertension patients versus healthy subjects. The metabolic phenotype of Uygur hypertension patients shows significant heteromorphosis, with the 12 characteristic metabolites as potential biomarkers of hypertension.